aye, that works fine. wanted to post it up for kfs + davo though 
In-Development TradeDangerous: power-user trade optimizer
- Thread starter kfsone
- Start date
factions?
Are there any plans to include the faction for each station and system?
Maybe just all 3 Superpowers (federation, empire and alliance) and Independent for all independent systems.
If you just want to trade, that missing will be no problem, but if you try to use force against one nation, you'll get problems using tradedangerous while your traderuns. (I'd guess, never tried force before)
Are there any plans to include the faction for each station and system?
Maybe just all 3 Superpowers (federation, empire and alliance) and Independent for all independent systems.
If you just want to trade, that missing will be no problem, but if you try to use force against one nation, you'll get problems using tradedangerous while your traderuns. (I'd guess, never tried force before)
Ah, that is a big stock value - I will fix the code to handle that. The columns will be a little out on that line.
OK fixed. Try now.
C:\Temp\Trade_Dangerous>trade.py import --plug=maddavo
NOTE: Stale/missing local copy, downloading full .prices file.
Connecting to server: http://www.davek.com.au/td/prices.asp
PLUGIN ERROR: Data is not Maddavo's prices list:
C:\Temp\Trade_Dangerous>
Odd since I do have a working .prices file having edited the bad stock data to a sensible number
NOTE: Stale/missing local copy, downloading full .prices file.
Connecting to server: http://www.davek.com.au/td/prices.asp
PLUGIN ERROR: Data is not Maddavo's prices list:
C:\Temp\Trade_Dangerous>
Odd since I do have a working .prices file having edited the bad stock data to a sensible number
I have discovered a (possibly odd) behaviour
In the general style
trade.py run --ca 10 --cr 1000 --from start/stn --to end/stn --hops 1 --ly 10
based in the time it takes to return a result, it seems to be doing a lot of thinking about route, even if the distance from start to end is less than the range given.
For example, if I put --ly=100 (I can't remember what my range is, but it's enough and please tell me the best trade without me having to alt-tab again) it will take a LOT longer to work out than it's --ly=16.89 (okok... I alt-tabbed are you happy now).
Is there any reason for doing some long calculation to route from A->B where LY > distance(A->B) ??
In the general style
trade.py run --ca 10 --cr 1000 --from start/stn --to end/stn --hops 1 --ly 10
based in the time it takes to return a result, it seems to be doing a lot of thinking about route, even if the distance from start to end is less than the range given.
For example, if I put --ly=100 (I can't remember what my range is, but it's enough and please tell me the best trade without me having to alt-tab again) it will take a LOT longer to work out than it's --ly=16.89 (okok... I alt-tabbed are you happy now).
Is there any reason for doing some long calculation to route from A->B where LY > distance(A->B) ??
Last edited:
The est xxx cr total at the end of a trade run needs to add up with the optional insurance btw.
If you're using the current main branch, then mostly just loading in all of the possible trades for all of the stations in a 100 ly^3 volume around your origin.
But even with the experimental branch it'll still be slow because I haven't special cased the hop-to-destination case so it has to load all the stations in a volume of r=100ly; it's literally looking through the volume at ALL the stations and asking "is this one of them?"
My apologies if this has come up already (70 pages of posts turned out to be a bit much to go through in detail), but two quick questions:
1) Is there a way to search for circular routes. Using the same --to and --from works of course, but if I want to find the best circular route in 10 systems' radius for example?
2) Is there any ETA on including pad size as a search parameter for the "run" command? (Just got my Type-9 (yay!) and it's a pain to grind through all route suggestions weeding out outposts)
1) Is there a way to search for circular routes. Using the same --to and --from works of course, but if I want to find the best circular route in 10 systems' radius for example?
2) Is there any ETA on including pad size as a search parameter for the "run" command? (Just got my Type-9 (yay!) and it's a pain to grind through all route suggestions weeding out outposts)
1) Yeah, to and from set the same and -hops 10 set should do the job, with range setting
2) From an earlier post by kfsone, very soon (maybe days away)
On 2), thanks for the update
On 1) my point was to find the best roundtrip between two stations without already knowing which station to start with. See what I mean? I'd just like to find the best circular route in my vicinity, without having to try all the stations, one by one, using the same --to and --from.
On 2), thanks for the update
On 1) my point was to find the best roundtrip between two stations without already knowing which station to start with. See what I mean? I'd just like to find the best circular route in my vicinity, without having to try all the stations, one by one, using the same --to and --from.
Given the way that the database has grown in the past few weeks and the way the search function is currently written it's more and more unlikely a search of that type will be effective going forward. Unless of course someone comes up with a significantly better way to deal with filtering good-candidate stations without loading ridiculous amounts of nodes into a matrix in memory.
Right now, I think kfsone is working on culling the matrix of stations based on proximity which should be a very good start to solving the issues with known start/end points. I already tagged something similar to your issue on the bitbucket tracker.
Turns out I installed 32 bit Python 3.4.2 thinking I was upgrading 64 bit Python 3.4.1. It was a real pain to get both of these uninstalled/cleaned off.
A reinstall of Python 3.4.2 64 bit resolved my error.
Thank you !
Ok, thanks! I'm always trying to get the best possible roundtrip, rather than just going into a random direction based on the last good route, and have in the past been able to find some awesome "triangle routes", but only with medium pads. With a Type-9 those suddenly became useless, and I'm struggling to come up with new ones. But for now I'll try patiently to test different origin points manually.
Quick update:
Max Pad Size for Run/Buy/Sell/Local/Rare has been added
Grab the latest code (http://kfs.org/td/source) and use the issue tracker to report any problems (http://kfs.org/td/issues)
The argument is "--pad-size" which abbreviates to "--pad" or "-p" and it takes a list of the pad sizes you'll accept:
- - - - - Additional Content Posted / Auto Merge - - - - -
it sounds like you're trying to be a more involved participant in your trading, which is excellent, but it's VERY rare that round-trips are efficient, FDev put a lot of work into the economy so this would be true.
However, you can use the "-s" argument to run to start from roughly the area you are in. E.g. if you're at sol and you'd like to know what's hot within a couple of systems, try:
(-s is short for --start-jumps and it takes an argument which is a number of jumps away from your start location to try additional stations as a start location)
This will try routes from all the stations in sol and all the stations in systems you could reach with 3 jumps (-s3/--start-jumps 3/--start-jumps=3/--start 3)
But you might want to instead try "--hops 3" and "--routes 10" and then take a look through what comes up.
- - - - - Additional Content Posted / Auto Merge - - - - -
Just committed it - one of our users contributed an initial stab at implement this for run, I went ahead and fleshed it out and added it to run/local/buy/sell/rares.
The out-of-memory/performance fix is in testing and I won't have time tonight to make it live.
Max Pad Size for Run/Buy/Sell/Local/Rare has been added
Grab the latest code (http://kfs.org/td/source) and use the issue tracker to report any problems (http://kfs.org/td/issues)
The argument is "--pad-size" which abbreviates to "--pad" or "-p" and it takes a list of the pad sizes you'll accept:
Code:
--pad s small only
--pad sml? any pad size (same as not using --pad)
--pad s? small or unknown
--pad SsMmLl? just being silly
--pad l? large or unknown
--pad ? *only* unknown*- - - - - Additional Content Posted / Auto Merge - - - - -
it sounds like you're trying to be a more involved participant in your trading, which is excellent, but it's VERY rare that round-trips are efficient, FDev put a lot of work into the economy so this would be true.
However, you can use the "-s" argument to run to start from roughly the area you are in. E.g. if you're at sol and you'd like to know what's hot within a couple of systems, try:
(-s is short for --start-jumps and it takes an argument which is a number of jumps away from your start location to try additional stations as a start location)
Code:
$ trade.py run --from sol --start-jumps 3 ... --hops 2
aka
$ trade.py run --from sol -s3 --hops 2This will try routes from all the stations in sol and all the stations in systems you could reach with 3 jumps (-s3/--start-jumps 3/--start-jumps=3/--start 3)
But you might want to instead try "--hops 3" and "--routes 10" and then take a look through what comes up.
- - - - - Additional Content Posted / Auto Merge - - - - -
Just committed it - one of our users contributed an initial stab at implement this for run, I went ahead and fleshed it out and added it to run/local/buy/sell/rares.
Code:
Station Cost DistLy Age/days StnLs B/mkt Pad
--------------------------------------------------------------------------------------
GROOMBRIDGE 1618/Gernsback Terminal 250 0.00 0.74 ? ? ?
GROOMBRIDGE 1618/Franklin Ring 228 0.00 14.76 ? ? ?
GROOMBRIDGE 1618/Cady Market 38 0.00 15.31 169 Yes Lrg
GROOMBRIDGE 1618/Martinez Market 38 0.00 15.32 ? ? ?
DUAMTA/Davis Terminal 39 7.88 10.56 550 Yes Lrg
WISE 1506+7027/Dobrovolskiy Enterprise 255 10.18 5.62 249 Yes Lrg
ROSS 104/Bunch City 143 10.79 15.21 ? ? ?
...The out-of-memory/performance fix is in testing and I won't have time tonight to make it live.
Last edited:
Just confirm in case I missed something, all outpost stations are s + m only, all 'slot' stations have l pads? Since a lot of '?' can be filled in via patient system view on the map I'd like to know if there are any odd cases before committing bad data.
As far as i know that is correct.
@kfsone
What is the ShipVendor.csv used for? Ok, obviously it is for storing where which ship is available, but is there something like "trade buy Anaconda --near Eranin"? Or is adding data to the ShipVendor.csv more or less useless at the moment?
If it is not useless, an add command would be great, for example like "trade.py shipvendor --add "i bootis/nowhere port" -ships anaconda,viper,cobra,..."
Confirm that this concurs with my experience. I've not seen any other configurations.
Confirmed for my experience.
I figure we'll probably do something like that eventually, in the meantime some people were already starting to collect some of the data anyway. And a batch-add way would definitely be easier than editing the .csv file
I hate to me too, but me too.
And if we're wrong, the difference is that we had 'M' instead of '?'.
I'm flying a T6 and I've not found a station I wasn't able to dock at. That's not to say there haven't been instances of being denied docking permission, but that's usually been because something was using the 'M' pad or the instance I was in was bugged and relogging cleared it.
I half-suspect that much of the "S" fear comes from people who didn't realize that the game simulates ATC for AI traffic so there are times even in a solo instance where you can't dock because the station is "busy".